Estimation of minimal maximal absolute residual of non-linear model;
最小最大绝对残差方法能保证最大绝对残差下降 ,这是非线性模型参数估计计算收敛的重要原因 ,算例表明最小二乘法比最小最大绝对残差方法对初值敏感 。
In this paper, a new genetic algorithm based on nonlinear least absolute correction sum is designed.
文章在扼要介绍遗传算法的基础上,设计了非线性残差绝对值和最小的遗传算法,并将其运用于非线性残差绝对值和最小的平差模型中,并通过实例验证了该算法的有效性。
Generally, the application of the adjustment model of minimizing the sum of absolute residuals in satellitic linear array scanner imagery resection not only can overcome the strong relativity of exterior orientation elements, but also can eliminate gross error effectively.
利用残差绝对值和最小法进行单线阵CCD卫星影像的后方交会 ,一般来说 ,不仅可以克服外方位元素之间的强相关性 ,而且可以有效地剔除粗差。